Exploring emotions on wine websites : finding joy

Abstract

There is little research on the emotions expressed by wine estate websites and whether the emotions conveyed are related to or can significantly predict the positive or negative sentiment articulated by these websites. In order to examine the emotions and sentiment of wine estate websites, content from the websites of 10 wine estates from 5 countries well known for their diverse and high-quality wines (South Africa, Australia, New Zealand, France, the USA) were selected and analyzed using IBM鈥檚 Watson. IBM鈥檚 Watson assessed the emotions and sentiment expressed on each website, and an output was then used in a multiple regression analysis with emotions predicting sentiment. The findings show that emotions are not related significantly to the overall sentiment on the winery websites. Finally, this research shows how wine estates and wine marketers can test whether the content of their websites mirrors the emotions that they desire to express to consumers and visitors, which can assure that appropriate expectations are set for the consumption experience.